Releasesby GlyphUncached Rate Limits Cerebras now uses a dual-bucket rate limiting model
Uncached Rate Limits Cerebras now uses a dual-bucket rate limiting model. Every organization has two independent token limits: Uncached TPM , tokens that require full compute (cache misses). Total TPM , uncached + cached tokens combined.
Also shipped Jul 16, 2026Cerebras in July 2026 →
DCM Projects: New capabilities (Preview)
DCM Projects now supports the following capabilities in preview: For more information, see Supported object types in DCM Projects and Deploy and manage DCM Projects .
VOLATILE scalar UDFs are supported with dynamic table incremental refresh (General availability)
VOLATILE scalar user-defined functions (UDFs) are now supported with dynamic table incremental refresh in INCREMENTAL, AUTO, and ADAPTIVE refresh modes. This applies to scalar UDFs written in Python, Java, Scala, JavaScript, and SQL. For incremental refresh, VOLATILE UDFs are supported only in the SELECT clause.
Scala and Java UDFs in Unity Catalog are now generally available
Register Scala and Java user-defined functions (UDFs) in Unity Catalog and reuse them across notebooks, jobs, and SQL warehouses with Unity Catalog governance. Reuse existing JVM logic and get better performance than Python UDFs on all compute, including serverless. See Scala and Java user-defined functions (UDFs) in Unity Catalog .
MLflow trace storage in Unity Catalog is now generally available
MLflow trace storage in Unity Catalog tables is now generally available. Databricks recommends storing traces in Unity Catalog for new and production workloads: storage is unlimited, traces are stored in OpenTelemetry (OTel) format, access is governed through Unity Catalog schema and table permissions, and you can query traces with SQL, Genie, AI/BI...
Agentic code converter is in Beta
The agentic code converter is in Beta . Use it to convert legacy SQL dialects, such as T-SQL, Snowflake, Redshift, Oracle, BigQuery, and Teradata, to ANSI SQL in migration projects in your workspace.
Google Gemini 3 Pro is retired
Google Gemini 3 Pro is retired. Gemini 3.1 Pro is the recommended replacement. See Deprecated and retired models for migration guidance.
More from Cerebras InferenceEvery Cerebras Inference release →
Free Trial credits for new accounts New accounts now receive \$5 in free credits after adding a verified...
Free Trial credits for new accounts New accounts now receive \$5 in free credits after adding a verified payment method. Credits work across all models on public endpoints and expire 30 days after Cerebras grants them. You aren't charged unless you purchase additional credits.
Gemma 4 31B available across coding integrations gemma-4-31b is now a selectable model in our Cline, Kilo...
Gemma 4 31B available across coding integrations gemma-4-31b is now a selectable model in our Cline, Kilo Code, opencode, and VS Code integration guides. Gemma 4 31B adds multimodal support to coding workflows , you can pass screenshots, diagrams, and design assets alongside text prompts.
API Version 2 Default Rollout As of July 22, 2026, API version 2 is the default
API Version 2 Default Rollout As of July 22, 2026, API version 2 is the default. Version 2 includes: Stricter schema validation for structured outputs Stricter validation for multi-turn tool-calling messages A separate reasoning_logprobs response field for reasoning-token logprobs Corrected Unicode handling in logprobs Z.ai GLM 4.7 is scheduled for...
Image limit increased for Developer and Enterprise accounts Developer and Enterprise accounts using...
Image limit increased for Developer and Enterprise accounts Developer and Enterprise accounts using gemma-4-31b on public endpoints can now include up to 10 images in a single request, increased from 5. Free Trial accounts remain limited to 2 images per request.
Predicted Outputs limited to dedicated endpoints Predicted Outputs is available by request on dedicated...
Predicted Outputs limited to dedicated endpoints Predicted Outputs is available by request on dedicated endpoints. To request it for your endpoint, contact us or reach out to your account representative. You can also try Predicted Outputs with gpt-oss-120b in Public Preview using a Free Trial or Developer account.
Image dimension limit for image inputs Image inputs now enforce a maximum width and height of 15,000 pixels...
Image dimension limit for image inputs Image inputs now enforce a maximum width and height of 15,000 pixels per side. Requests with larger images return 413 Content Too Large with the message "Image dimensions {width}x{height} exceed the 15000px per-side limit". Existing payload-size, image-count, and decompressed-memory limits are unchanged.
GLM 4.7 reasoning_logprobs default will not ship before deprecation zai-glm-4.7 is scheduled for deprecation...
GLM 4.7 reasoning_logprobs default will not ship before deprecation zai-glm-4.7 is scheduled for deprecation on August 17, 2026 and will keep its existing default logprobs response shape until then. Requests that explicitly set the X-Cerebras-Version-Patch: 2 header still receive the separated reasoning_logprobs field.
Gemma 4 31B and Image Input Support - Gemma 4 31B (gemma-4-31b) is now available in preview
Gemma 4 31B and Image Input Support Gemma 4 31B (gemma-4-31b) is now available in preview. It's the first model on Cerebras to support image input. Send screenshots, documents, diagrams, and design assets alongside text using the standard Chat Completions API. Image Inputs are now available in Public Preview through the Chat Completions API.